SPs, Life and the Universe

A while back, a fair amount of dissatisfaction with the SP system had been expressed in this thread. I appreciate those who those who have proposed reasonable alternatives, including TRSPs and the sum of winning and losing points. Given that the name SP is intended to represent “strength of schedule points”, a system that truly represents strength of schedule as accurately as possible is desirable.

While I don’t have an issue with changing this particular rule, I must admit that I have some discomfort with the direction that a rule change like this (and other trends) could bring. Until the 2009 season, SP’s were originally called RP’s (rank points), and were simply a means of ranking teams when multiple teams had the same number of QP’s. The original game descriptions never explicitly stated why the losing alliance’s points were chosen as a tie-breaker, but teams inferred that it was a way to reward alliances that beat high-scoring teams (implying a stronger win), but also to encourage offense over defense, which gives a psychological boost to lower-level teams (“we didn’t win, but at least we scored decently”). While using the difference between winner’s and loser’s points would have been more representative of a team’s true strength, the idea of sacrificing “truer representation” for “bringing up the bottom” was one that seemed to be acceptable, based on the lack of complaints during those early years (2005-2008).

While using random means to narrow the gap between the top and bottom performers may not seem sensible, most enjoyable games do this all the time. There are games of pure skill like chess, which very few players have the skill to enjoy. In contrast, “lame” kids games (like Candyland), are pure randomness and can be played by anyone, but honestly aren’t fun. The “best” games (like Risk or card games) include some element of chance (dice, cards), but heavily favor those who have acquired a certain level of skill. These games appeal to a larger segment of the population than chess, because those who make a lesser investment can still enjoy the game, but it’s clear that those who “advance to the big leagues” are those who invest the time to become “experts”.

VRC’s Driver Skills and Programming Skillls Challenges are “pure skill”. In contrast, the VRC game’s primary randomizer is the alliance system, and the scouting process already compensates for the randomness to some degree. Good teams have employed strategies to further compensate for the uncertainties caused by randomness, like farming SP’s and protecting goals. The $64K question is, “Should the VRC Game also trend toward ‘pure skill’ with randomization reduced or even eliminated?”

A “yes” is more beneficial to high-level teams, and a “no” is more appealing to low-level ones. While this doesn’t necessarily make a “yes” answer wrong (no one wants a lame, totally random game), my concern is that disproportionate attention to the concerns of high-level teams while ignoring lower-level ones will push a certain demographic out of the Vex community. Some of my Vex teams have been “home” to some unlikely students: kids with disabilities, kids who go hungry, kids who never had a close friend or family member graduate from high school. And some of those kids have been accepted to 4-year universities. Unfortunately, each year, it gets harder to recruit and keep students because it’s harder to “stay with the pack”.

Perhaps it would be reasonable to have one set of rules for regional events and slightly different ones for Nationals and Worlds. Ultimately, a program needs to serve the students it has, and the teams that advance will probably have different needs the ones that don’t.

There may come a time when I don’t have a team because teams like mine can’t keep up with the pace, but I think something of value will be lost when that time comes.

2 Likes

While I wish I had some easy answers to the questions you raise, and scenarios you describe – I don’t. As you mentioned (and was mentioned in the thread you linked) there are a lot of considerations in play in the decision making process of the GDC.

We ARE listening (and reading). Your comments are not falling on deaf ears, and the interests of all teams (including those like the ones you describe) are taken into account.

One question – do you feel like you’re being left behind at the local level? The state level? The world level?

Feel free to email me if you’d like to discuss further: jvn@vexrobotics.com

-John

1 Like

In answer to your question, it’s primarily at the local one, as we don’t have expectations of advancement.

Thanks, John, for considering this matter. I’ll e-mail you within the next few days.

Yolande

1 Like

I’ve also followed these comments and have many of the same reservations as the original thread. I can appreciate different rules at different levels though. In our local league we have kids that struggle to get any sacks scored right next to a team that can score 150+ points.

The stronger team ends up scoring 50+ points for their opposition and 75+ for themselves. To what end I’m not sure because it is very clear which team is superior and I’m not sure how good a team feels when their opposition is scoring on their behalf just to better their own positioning.

Perhaps for next year and beyond a more elaborate handicapping system can be devised that truly represents strength of schedule. The mechanism to do this could be any number of different algorithms, no new to reinvent the wheel here.

It becomes an RPI for robotics if you will. Having more regional/state qualifiers and doing them earlier and more often would help establish this. The teams that have real aspirations of going to Nationals and Worlds could go to these large events to establish their ranking. At the end of the season the top 25 SoS teams that have not already earned a world ranking get invited.

1 Like

It’s clear that the SP system could be improved. I think my discomfort with revamping the SP’s is that over the years, a number of changes have been made, and most of them favor high-scoring teams. A mental inventory:

Changes that have greater benefit to high-scoring teams

  • More matches (increased from 3 in 2007 to 6 in 2012 at local events)
  • Less attention to interviews and judged awards (because of more matches)
  • Better, stronger, more expensive parts
  • Games with higher goals
  • Reduction of match time from 2:20 to 2:00 (total)
  • Removal of a no-lift scoring opportunity from the 2010 game, Roundup

Changes that have greater benefit to low-scoring teams

  • Reinstatement of no-lift scoring to 2011 game, Gateway (thanks, GDC)
  • Repositioning rule during autonomous

Each team, whether high or low-level, is responsible for figuring out how to use the system to its best advantage, and “complaining nicely” to the right people can be one appropriate and effective way of doing this. My concern is that overuse of this single strategy can short-circuit the process of problem-solving in multiple ways. A nimble team is able to say, “This rule isn’t ideal and it may not be doing what is intended, but here’s how we can use it to our best advantage.” Despite the changes listed above, even my low-scoring team has found “workarounds” to show itself in its best light, and teams at the top ought to be even more capable of this.

Frankly, Vex is one of the most adaptable businesses I’ve ever worked with, and students who expect the companies they work for to be as responsive to complaints are in for an unpleasant surprise. One of my former students (now working) told me, “I like 80% of my job, but I eat dirt 20% of the time and that’s not going to change. I’m still better off than the people who hate 100% of their jobs but do it for a paycheck.”

I think a change in the SP system would ultimately be more fair, but finding ways to reward teams that excel in other ways than scoring is also important. One thought (which I’ve submitted in more detail) is to increase judged awards at the local level and have 2 categories of advancement: tournament status (tournament play as usual) and exhibition status (for judged awards, like the design award or winners of online challenges). On one hand, a greater variety of robots would be likely to be on display, but high-scoring teams would not be frustrated with a low-scoring alliance partner and be tempted to think, “They don’t belong here.”

2 Likes